I am trying to do a calculation on a supercell of rutile TiO2 with an oxygen vacancy. I am pre-converging the geometry with VASP and building the FLEUR input file from VASP's converged CONTCAR. When I run this job FLEUR exits during initialisation with the error:
**************FLEUR-Error*****************
Error message: symmetry is incorrect for some atomp
Error occurred in subroutine: ptsym
Error from PE:0/1

My input file and inp.xml are attached.

I expect VASP to be maintaining the input symmetry. Nonetheless I used sglib (via phonopy) to ensure I was running with a SG-symmetric geometry. Cm or Amm2 symmetry can be identified (depending on the tolerance) and running with either the conventional or primitive symmetrised geometries gave the same error.

If I run inpgen with the -nosym flag I do not get the error.

This is all with MaX 8.0.

Any hints about what's happening with the symmetry in FLEUR?

Yep, that solved the issue. Interestingly six is the right number. If I give seven significant digits in the coordinates I get the wrong symmetry error.

Thanks for the rapid response.